Hide keyboard shortcuts

Hot-keys on this page

r m x p   toggle line displays

j k   next/prev highlighted chunk

0   (zero) top of page

1   (one) first highlighted chunk

1# This file is part of daf_butler. 

2# 

3# Developed for the LSST Data Management System. 

4# This product includes software developed by the LSST Project 

5# (http://www.lsst.org). 

6# See the COPYRIGHT file at the top-level directory of this distribution 

7# for details of code ownership. 

8# 

9# This program is free software: you can redistribute it and/or modify 

10# it under the terms of the GNU General Public License as published by 

11# the Free Software Foundation, either version 3 of the License, or 

12# (at your option) any later version. 

13# 

14# This program is distributed in the hope that it will be useful, 

15# but WITHOUT ANY WARRANTY; without even the implied warranty of 

16# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 

17# GNU General Public License for more details. 

18# 

19# You should have received a copy of the GNU General Public License 

20# along with this program. If not, see <http://www.gnu.org/licenses/>. 

21 

22""" 

23Butler top level classes. 

24""" 

25from __future__ import annotations 

26 

27__all__ = ( 

28 "Butler", 

29 "ButlerValidationError", 

30 "PruneCollectionsArgsError", 

31 "PurgeWithoutUnstorePruneCollectionsError", 

32 "RunWithoutPurgePruneCollectionsError", 

33 "PurgeUnsupportedPruneCollectionsError", 

34) 

35 

36 

37from collections import defaultdict 

38import contextlib 

39import logging 

40import numbers 

41import os 

42from typing import ( 

43 Any, 

44 ClassVar, 

45 Counter, 

46 Dict, 

47 Iterable, 

48 Iterator, 

49 List, 

50 MutableMapping, 

51 Optional, 

52 Set, 

53 TextIO, 

54 Tuple, 

55 Type, 

56 Union, 

57) 

58 

59try: 

60 import boto3 

61except ImportError: 

62 boto3 = None 

63 

64from lsst.utils import doImport 

65from .core import ( 

66 AmbiguousDatasetError, 

67 ButlerURI, 

68 Config, 

69 ConfigSubset, 

70 DataCoordinate, 

71 DataId, 

72 DataIdValue, 

73 DatasetRef, 

74 DatasetType, 

75 Datastore, 

76 Dimension, 

77 DimensionConfig, 

78 FileDataset, 

79 Progress, 

80 StorageClassFactory, 

81 Timespan, 

82 ValidationError, 

83) 

84from .core.repoRelocation import BUTLER_ROOT_TAG 

85from .core.utils import transactional, getClassOf 

86from ._deferredDatasetHandle import DeferredDatasetHandle 

87from ._butlerConfig import ButlerConfig 

88from .registry import ( 

89 Registry, 

90 RegistryConfig, 

91 RegistryDefaults, 

92 CollectionSearch, 

93 CollectionType, 

94 ConflictingDefinitionError, 

95 DatasetIdGenEnum, 

96) 

97from .transfers import RepoExportContext 

98 

99log = logging.getLogger(__name__) 

100 

101 

102class ButlerValidationError(ValidationError): 

103 """There is a problem with the Butler configuration.""" 

104 pass 

105 

106 

107class PruneCollectionsArgsError(TypeError): 

108 """Base class for errors relating to Butler.pruneCollections input 

109 arguments. 

110 """ 

111 pass 

112 

113 

114class PurgeWithoutUnstorePruneCollectionsError(PruneCollectionsArgsError): 

115 """Raised when purge and unstore are both required to be True, and 

116 purge is True but unstore is False. 

117 """ 

118 

119 def __init__(self) -> None: 

120 super().__init__("Cannot pass purge=True without unstore=True.") 

121 

122 

123class RunWithoutPurgePruneCollectionsError(PruneCollectionsArgsError): 

124 """Raised when pruning a RUN collection but purge is False.""" 

125 

126 def __init__(self, collectionType: CollectionType): 

127 self.collectionType = collectionType 

128 super().__init__(f"Cannot prune RUN collection {self.collectionType.name} without purge=True.") 

129 

130 

131class PurgeUnsupportedPruneCollectionsError(PruneCollectionsArgsError): 

132 """Raised when purge is True but is not supported for the given 

133 collection.""" 

134 

135 def __init__(self, collectionType: CollectionType): 

136 self.collectionType = collectionType 

137 super().__init__( 

138 f"Cannot prune {self.collectionType} collection {self.collectionType.name} with purge=True.") 

139 

140 

141class Butler: 

142 """Main entry point for the data access system. 

143 

144 Parameters 

145 ---------- 

146 config : `ButlerConfig`, `Config` or `str`, optional. 

147 Configuration. Anything acceptable to the 

148 `ButlerConfig` constructor. If a directory path 

149 is given the configuration will be read from a ``butler.yaml`` file in 

150 that location. If `None` is given default values will be used. 

151 butler : `Butler`, optional. 

152 If provided, construct a new Butler that uses the same registry and 

153 datastore as the given one, but with the given collection and run. 

154 Incompatible with the ``config``, ``searchPaths``, and ``writeable`` 

155 arguments. 

156 collections : `str` or `Iterable` [ `str` ], optional 

157 An expression specifying the collections to be searched (in order) when 

158 reading datasets. 

159 This may be a `str` collection name or an iterable thereof. 

160 See :ref:`daf_butler_collection_expressions` for more information. 

161 These collections are not registered automatically and must be 

162 manually registered before they are used by any method, but they may be 

163 manually registered after the `Butler` is initialized. 

164 run : `str`, optional 

165 Name of the `~CollectionType.RUN` collection new datasets should be 

166 inserted into. If ``collections`` is `None` and ``run`` is not `None`, 

167 ``collections`` will be set to ``[run]``. If not `None`, this 

168 collection will automatically be registered. If this is not set (and 

169 ``writeable`` is not set either), a read-only butler will be created. 

170 searchPaths : `list` of `str`, optional 

171 Directory paths to search when calculating the full Butler 

172 configuration. Not used if the supplied config is already a 

173 `ButlerConfig`. 

174 writeable : `bool`, optional 

175 Explicitly sets whether the butler supports write operations. If not 

176 provided, a read-write butler is created if any of ``run``, ``tags``, 

177 or ``chains`` is non-empty. 

178 inferDefaults : `bool`, optional 

179 If `True` (default) infer default data ID values from the values 

180 present in the datasets in ``collections``: if all collections have the 

181 same value (or no value) for a governor dimension, that value will be 

182 the default for that dimension. Nonexistent collections are ignored. 

183 If a default value is provided explicitly for a governor dimension via 

184 ``**kwargs``, no default will be inferred for that dimension. 

185 **kwargs : `str` 

186 Default data ID key-value pairs. These may only identify "governor" 

187 dimensions like ``instrument`` and ``skymap``. 

188 

189 Examples 

190 -------- 

191 While there are many ways to control exactly how a `Butler` interacts with 

192 the collections in its `Registry`, the most common cases are still simple. 

193 

194 For a read-only `Butler` that searches one collection, do:: 

195 

196 butler = Butler("/path/to/repo", collections=["u/alice/DM-50000"]) 

197 

198 For a read-write `Butler` that writes to and reads from a 

199 `~CollectionType.RUN` collection:: 

200 

201 butler = Butler("/path/to/repo", run="u/alice/DM-50000/a") 

202 

203 The `Butler` passed to a ``PipelineTask`` is often much more complex, 

204 because we want to write to one `~CollectionType.RUN` collection but read 

205 from several others (as well):: 

206 

207 butler = Butler("/path/to/repo", run="u/alice/DM-50000/a", 

208 collections=["u/alice/DM-50000/a", 

209 "u/bob/DM-49998", 

210 "HSC/defaults"]) 

211 

212 This butler will `put` new datasets to the run ``u/alice/DM-50000/a``. 

213 Datasets will be read first from that run (since it appears first in the 

214 chain), and then from ``u/bob/DM-49998`` and finally ``HSC/defaults``. 

215 

216 Finally, one can always create a `Butler` with no collections:: 

217 

218 butler = Butler("/path/to/repo", writeable=True) 

219 

220 This can be extremely useful when you just want to use ``butler.registry``, 

221 e.g. for inserting dimension data or managing collections, or when the 

222 collections you want to use with the butler are not consistent. 

223 Passing ``writeable`` explicitly here is only necessary if you want to be 

224 able to make changes to the repo - usually the value for ``writeable`` can 

225 be guessed from the collection arguments provided, but it defaults to 

226 `False` when there are not collection arguments. 

227 """ 

228 def __init__(self, config: Union[Config, str, None] = None, *, 

229 butler: Optional[Butler] = None, 

230 collections: Any = None, 

231 run: Optional[str] = None, 

232 searchPaths: Optional[List[str]] = None, 

233 writeable: Optional[bool] = None, 

234 inferDefaults: bool = True, 

235 **kwargs: str, 

236 ): 

237 defaults = RegistryDefaults(collections=collections, run=run, infer=inferDefaults, **kwargs) 

238 # Load registry, datastore, etc. from config or existing butler. 

239 if butler is not None: 

240 if config is not None or searchPaths is not None or writeable is not None: 

241 raise TypeError("Cannot pass 'config', 'searchPaths', or 'writeable' " 

242 "arguments with 'butler' argument.") 

243 self.registry = butler.registry.copy(defaults) 

244 self.datastore = butler.datastore 

245 self.storageClasses = butler.storageClasses 

246 self._config: ButlerConfig = butler._config 

247 else: 

248 self._config = ButlerConfig(config, searchPaths=searchPaths) 

249 if "root" in self._config: 

250 butlerRoot = self._config["root"] 

251 else: 

252 butlerRoot = self._config.configDir 

253 if writeable is None: 

254 writeable = run is not None 

255 self.registry = Registry.fromConfig(self._config, butlerRoot=butlerRoot, writeable=writeable, 

256 defaults=defaults) 

257 self.datastore = Datastore.fromConfig(self._config, self.registry.getDatastoreBridgeManager(), 

258 butlerRoot=butlerRoot) 

259 self.storageClasses = StorageClassFactory() 

260 self.storageClasses.addFromConfig(self._config) 

261 if "run" in self._config or "collection" in self._config: 

262 raise ValueError("Passing a run or collection via configuration is no longer supported.") 

263 

264 GENERATION: ClassVar[int] = 3 

265 """This is a Generation 3 Butler. 

266 

267 This attribute may be removed in the future, once the Generation 2 Butler 

268 interface has been fully retired; it should only be used in transitional 

269 code. 

270 """ 

271 

272 @staticmethod 

273 def makeRepo(root: str, config: Union[Config, str, None] = None, 

274 dimensionConfig: Union[Config, str, None] = None, standalone: bool = False, 

275 searchPaths: Optional[List[str]] = None, forceConfigRoot: bool = True, 

276 outfile: Optional[str] = None, overwrite: bool = False) -> Config: 

277 """Create an empty data repository by adding a butler.yaml config 

278 to a repository root directory. 

279 

280 Parameters 

281 ---------- 

282 root : `str` or `ButlerURI` 

283 Path or URI to the root location of the new repository. Will be 

284 created if it does not exist. 

285 config : `Config` or `str`, optional 

286 Configuration to write to the repository, after setting any 

287 root-dependent Registry or Datastore config options. Can not 

288 be a `ButlerConfig` or a `ConfigSubset`. If `None`, default 

289 configuration will be used. Root-dependent config options 

290 specified in this config are overwritten if ``forceConfigRoot`` 

291 is `True`. 

292 dimensionConfig : `Config` or `str`, optional 

293 Configuration for dimensions, will be used to initialize registry 

294 database. 

295 standalone : `bool` 

296 If True, write all expanded defaults, not just customized or 

297 repository-specific settings. 

298 This (mostly) decouples the repository from the default 

299 configuration, insulating it from changes to the defaults (which 

300 may be good or bad, depending on the nature of the changes). 

301 Future *additions* to the defaults will still be picked up when 

302 initializing `Butlers` to repos created with ``standalone=True``. 

303 searchPaths : `list` of `str`, optional 

304 Directory paths to search when calculating the full butler 

305 configuration. 

306 forceConfigRoot : `bool`, optional 

307 If `False`, any values present in the supplied ``config`` that 

308 would normally be reset are not overridden and will appear 

309 directly in the output config. This allows non-standard overrides 

310 of the root directory for a datastore or registry to be given. 

311 If this parameter is `True` the values for ``root`` will be 

312 forced into the resulting config if appropriate. 

313 outfile : `str`, optional 

314 If not-`None`, the output configuration will be written to this 

315 location rather than into the repository itself. Can be a URI 

316 string. Can refer to a directory that will be used to write 

317 ``butler.yaml``. 

318 overwrite : `bool`, optional 

319 Create a new configuration file even if one already exists 

320 in the specified output location. Default is to raise 

321 an exception. 

322 

323 Returns 

324 ------- 

325 config : `Config` 

326 The updated `Config` instance written to the repo. 

327 

328 Raises 

329 ------ 

330 ValueError 

331 Raised if a ButlerConfig or ConfigSubset is passed instead of a 

332 regular Config (as these subclasses would make it impossible to 

333 support ``standalone=False``). 

334 FileExistsError 

335 Raised if the output config file already exists. 

336 os.error 

337 Raised if the directory does not exist, exists but is not a 

338 directory, or cannot be created. 

339 

340 Notes 

341 ----- 

342 Note that when ``standalone=False`` (the default), the configuration 

343 search path (see `ConfigSubset.defaultSearchPaths`) that was used to 

344 construct the repository should also be used to construct any Butlers 

345 to avoid configuration inconsistencies. 

346 """ 

347 if isinstance(config, (ButlerConfig, ConfigSubset)): 

348 raise ValueError("makeRepo must be passed a regular Config without defaults applied.") 

349 

350 # Ensure that the root of the repository exists or can be made 

351 uri = ButlerURI(root, forceDirectory=True) 

352 uri.mkdir() 

353 

354 config = Config(config) 

355 

356 # If we are creating a new repo from scratch with relative roots, 

357 # do not propagate an explicit root from the config file 

358 if "root" in config: 

359 del config["root"] 

360 

361 full = ButlerConfig(config, searchPaths=searchPaths) # this applies defaults 

362 datastoreClass: Type[Datastore] = doImport(full["datastore", "cls"]) 

363 datastoreClass.setConfigRoot(BUTLER_ROOT_TAG, config, full, overwrite=forceConfigRoot) 

364 

365 # if key exists in given config, parse it, otherwise parse the defaults 

366 # in the expanded config 

367 if config.get(("registry", "db")): 

368 registryConfig = RegistryConfig(config) 

369 else: 

370 registryConfig = RegistryConfig(full) 

371 defaultDatabaseUri = registryConfig.makeDefaultDatabaseUri(BUTLER_ROOT_TAG) 

372 if defaultDatabaseUri is not None: 

373 Config.updateParameters(RegistryConfig, config, full, 

374 toUpdate={"db": defaultDatabaseUri}, 

375 overwrite=forceConfigRoot) 

376 else: 

377 Config.updateParameters(RegistryConfig, config, full, toCopy=("db",), 

378 overwrite=forceConfigRoot) 

379 

380 if standalone: 

381 config.merge(full) 

382 else: 

383 # Always expand the registry.managers section into the per-repo 

384 # config, because after the database schema is created, it's not 

385 # allowed to change anymore. Note that in the standalone=True 

386 # branch, _everything_ in the config is expanded, so there's no 

387 # need to special case this. 

388 Config.updateParameters(RegistryConfig, config, full, toCopy=("managers",), overwrite=False) 

389 configURI: Union[str, ButlerURI] 

390 if outfile is not None: 

391 # When writing to a separate location we must include 

392 # the root of the butler repo in the config else it won't know 

393 # where to look. 

394 config["root"] = uri.geturl() 

395 configURI = outfile 

396 else: 

397 configURI = uri 

398 config.dumpToUri(configURI, overwrite=overwrite) 

399 

400 # Create Registry and populate tables 

401 registryConfig = RegistryConfig(config.get("registry")) 

402 dimensionConfig = DimensionConfig(dimensionConfig) 

403 Registry.createFromConfig(registryConfig, dimensionConfig=dimensionConfig, butlerRoot=root) 

404 

405 return config 

406 

407 @classmethod 

408 def _unpickle(cls, config: ButlerConfig, collections: Optional[CollectionSearch], run: Optional[str], 

409 defaultDataId: Dict[str, str], writeable: bool) -> Butler: 

410 """Callable used to unpickle a Butler. 

411 

412 We prefer not to use ``Butler.__init__`` directly so we can force some 

413 of its many arguments to be keyword-only (note that ``__reduce__`` 

414 can only invoke callables with positional arguments). 

415 

416 Parameters 

417 ---------- 

418 config : `ButlerConfig` 

419 Butler configuration, already coerced into a true `ButlerConfig` 

420 instance (and hence after any search paths for overrides have been 

421 utilized). 

422 collections : `CollectionSearch` 

423 Names of the default collections to read from. 

424 run : `str`, optional 

425 Name of the default `~CollectionType.RUN` collection to write to. 

426 defaultDataId : `dict` [ `str`, `str` ] 

427 Default data ID values. 

428 writeable : `bool` 

429 Whether the Butler should support write operations. 

430 

431 Returns 

432 ------- 

433 butler : `Butler` 

434 A new `Butler` instance. 

435 """ 

436 # MyPy doesn't recognize that the kwargs below are totally valid; it 

437 # seems to think '**defaultDataId* is a _positional_ argument! 

438 return cls(config=config, collections=collections, run=run, writeable=writeable, 

439 **defaultDataId) # type: ignore 

440 

441 def __reduce__(self) -> tuple: 

442 """Support pickling. 

443 """ 

444 return (Butler._unpickle, (self._config, self.collections, self.run, 

445 self.registry.defaults.dataId.byName(), 

446 self.registry.isWriteable())) 

447 

448 def __str__(self) -> str: 

449 return "Butler(collections={}, run={}, datastore='{}', registry='{}')".format( 

450 self.collections, self.run, self.datastore, self.registry) 

451 

452 def isWriteable(self) -> bool: 

453 """Return `True` if this `Butler` supports write operations. 

454 """ 

455 return self.registry.isWriteable() 

456 

457 @contextlib.contextmanager 

458 def transaction(self) -> Iterator[None]: 

459 """Context manager supporting `Butler` transactions. 

460 

461 Transactions can be nested. 

462 """ 

463 with self.registry.transaction(): 

464 with self.datastore.transaction(): 

465 yield 

466 

467 def _standardizeArgs(self, datasetRefOrType: Union[DatasetRef, DatasetType, str], 

468 dataId: Optional[DataId] = None, **kwds: Any 

469 ) -> Tuple[DatasetType, Optional[DataId]]: 

470 """Standardize the arguments passed to several Butler APIs. 

471 

472 Parameters 

473 ---------- 

474 datasetRefOrType : `DatasetRef`, `DatasetType`, or `str` 

475 When `DatasetRef` the `dataId` should be `None`. 

476 Otherwise the `DatasetType` or name thereof. 

477 dataId : `dict` or `DataCoordinate` 

478 A `dict` of `Dimension` link name, value pairs that label the 

479 `DatasetRef` within a Collection. When `None`, a `DatasetRef` 

480 should be provided as the second argument. 

481 kwds 

482 Additional keyword arguments used to augment or construct a 

483 `DataCoordinate`. See `DataCoordinate.standardize` 

484 parameters. 

485 

486 Returns 

487 ------- 

488 datasetType : `DatasetType` 

489 A `DatasetType` instance extracted from ``datasetRefOrType``. 

490 dataId : `dict` or `DataId`, optional 

491 Argument that can be used (along with ``kwds``) to construct a 

492 `DataId`. 

493 

494 Notes 

495 ----- 

496 Butler APIs that conceptually need a DatasetRef also allow passing a 

497 `DatasetType` (or the name of one) and a `DataId` (or a dict and 

498 keyword arguments that can be used to construct one) separately. This 

499 method accepts those arguments and always returns a true `DatasetType` 

500 and a `DataId` or `dict`. 

501 

502 Standardization of `dict` vs `DataId` is best handled by passing the 

503 returned ``dataId`` (and ``kwds``) to `Registry` APIs, which are 

504 generally similarly flexible. 

505 """ 

506 externalDatasetType: Optional[DatasetType] = None 

507 internalDatasetType: Optional[DatasetType] = None 

508 if isinstance(datasetRefOrType, DatasetRef): 

509 if dataId is not None or kwds: 

510 raise ValueError("DatasetRef given, cannot use dataId as well") 

511 externalDatasetType = datasetRefOrType.datasetType 

512 dataId = datasetRefOrType.dataId 

513 else: 

514 # Don't check whether DataId is provided, because Registry APIs 

515 # can usually construct a better error message when it wasn't. 

516 if isinstance(datasetRefOrType, DatasetType): 

517 externalDatasetType = datasetRefOrType 

518 else: 

519 internalDatasetType = self.registry.getDatasetType(datasetRefOrType) 

520 

521 # Check that they are self-consistent 

522 if externalDatasetType is not None: 

523 internalDatasetType = self.registry.getDatasetType(externalDatasetType.name) 

524 if externalDatasetType != internalDatasetType: 

525 raise ValueError(f"Supplied dataset type ({externalDatasetType}) inconsistent with " 

526 f"registry definition ({internalDatasetType})") 

527 

528 assert internalDatasetType is not None 

529 return internalDatasetType, dataId 

530 

531 def _findDatasetRef(self, datasetRefOrType: Union[DatasetRef, DatasetType, str], 

532 dataId: Optional[DataId] = None, *, 

533 collections: Any = None, 

534 allowUnresolved: bool = False, 

535 **kwds: Any) -> DatasetRef: 

536 """Shared logic for methods that start with a search for a dataset in 

537 the registry. 

538 

539 Parameters 

540 ---------- 

541 datasetRefOrType : `DatasetRef`, `DatasetType`, or `str` 

542 When `DatasetRef` the `dataId` should be `None`. 

543 Otherwise the `DatasetType` or name thereof. 

544 dataId : `dict` or `DataCoordinate`, optional 

545 A `dict` of `Dimension` link name, value pairs that label the 

546 `DatasetRef` within a Collection. When `None`, a `DatasetRef` 

547 should be provided as the first argument. 

548 collections : Any, optional 

549 Collections to be searched, overriding ``self.collections``. 

550 Can be any of the types supported by the ``collections`` argument 

551 to butler construction. 

552 allowUnresolved : `bool`, optional 

553 If `True`, return an unresolved `DatasetRef` if finding a resolved 

554 one in the `Registry` fails. Defaults to `False`. 

555 kwds 

556 Additional keyword arguments used to augment or construct a 

557 `DataId`. See `DataId` parameters. 

558 

559 Returns 

560 ------- 

561 ref : `DatasetRef` 

562 A reference to the dataset identified by the given arguments. 

563 

564 Raises 

565 ------ 

566 LookupError 

567 Raised if no matching dataset exists in the `Registry` (and 

568 ``allowUnresolved is False``). 

569 ValueError 

570 Raised if a resolved `DatasetRef` was passed as an input, but it 

571 differs from the one found in the registry. 

572 TypeError 

573 Raised if no collections were provided. 

574 """ 

575 datasetType, dataId = self._standardizeArgs(datasetRefOrType, dataId, **kwds) 

576 if isinstance(datasetRefOrType, DatasetRef): 

577 idNumber = datasetRefOrType.id 

578 else: 

579 idNumber = None 

580 timespan: Optional[Timespan] = None 

581 

582 # Process dimension records that are using record information 

583 # rather than ids 

584 newDataId: Dict[str, DataIdValue] = {} 

585 byRecord: Dict[str, Dict[str, Any]] = defaultdict(dict) 

586 

587 # if all the dataId comes from keyword parameters we do not need 

588 # to do anything here because they can't be of the form 

589 # exposure.obs_id because a "." is not allowed in a keyword parameter. 

590 if dataId: 

591 for k, v in dataId.items(): 

592 # If we have a Dimension we do not need to do anything 

593 # because it cannot be a compound key. 

594 if isinstance(k, str) and "." in k: 

595 # Someone is using a more human-readable dataId 

596 dimensionName, record = k.split(".", 1) 

597 byRecord[dimensionName][record] = v 

598 elif isinstance(k, Dimension): 

599 newDataId[k.name] = v 

600 else: 

601 newDataId[k] = v 

602 

603 # Go through the updated dataId and check the type in case someone is 

604 # using an alternate key. We have already filtered out the compound 

605 # keys dimensions.record format. 

606 not_dimensions = {} 

607 

608 # Will need to look in the dataId and the keyword arguments 

609 # and will remove them if they need to be fixed or are unrecognized. 

610 for dataIdDict in (newDataId, kwds): 

611 # Use a list so we can adjust the dict safely in the loop 

612 for dimensionName in list(dataIdDict): 

613 value = dataIdDict[dimensionName] 

614 try: 

615 dimension = self.registry.dimensions.getStaticDimensions()[dimensionName] 

616 except KeyError: 

617 # This is not a real dimension 

618 not_dimensions[dimensionName] = value 

619 del dataIdDict[dimensionName] 

620 continue 

621 

622 # Convert an integral type to an explicit int to simplify 

623 # comparisons here 

624 if isinstance(value, numbers.Integral): 

625 value = int(value) 

626 

627 if not isinstance(value, dimension.primaryKey.getPythonType()): 

628 for alternate in dimension.alternateKeys: 

629 if isinstance(value, alternate.getPythonType()): 

630 byRecord[dimensionName][alternate.name] = value 

631 del dataIdDict[dimensionName] 

632 log.debug("Converting dimension %s to %s.%s=%s", 

633 dimensionName, dimensionName, alternate.name, value) 

634 break 

635 else: 

636 log.warning("Type mismatch found for value '%r' provided for dimension %s. " 

637 "Could not find matching alternative (primary key has type %s) " 

638 "so attempting to use as-is.", 

639 value, dimensionName, dimension.primaryKey.getPythonType()) 

640 

641 # If we have some unrecognized dimensions we have to try to connect 

642 # them to records in other dimensions. This is made more complicated 

643 # by some dimensions having records with clashing names. A mitigation 

644 # is that we can tell by this point which dimensions are missing 

645 # for the DatasetType but this does not work for calibrations 

646 # where additional dimensions can be used to constrain the temporal 

647 # axis. 

648 if not_dimensions: 

649 # Calculate missing dimensions 

650 provided = set(newDataId) | set(kwds) | set(byRecord) 

651 missingDimensions = datasetType.dimensions.names - provided 

652 

653 # For calibrations we may well be needing temporal dimensions 

654 # so rather than always including all dimensions in the scan 

655 # restrict things a little. It is still possible for there 

656 # to be confusion over day_obs in visit vs exposure for example. 

657 # If we are not searching calibration collections things may 

658 # fail but they are going to fail anyway because of the 

659 # ambiguousness of the dataId... 

660 candidateDimensions: Set[str] = set() 

661 candidateDimensions.update(missingDimensions) 

662 if datasetType.isCalibration(): 

663 for dim in self.registry.dimensions.getStaticDimensions(): 

664 if dim.temporal: 

665 candidateDimensions.add(str(dim)) 

666 

667 # Look up table for the first association with a dimension 

668 guessedAssociation: Dict[str, Dict[str, Any]] = defaultdict(dict) 

669 

670 # Keep track of whether an item is associated with multiple 

671 # dimensions. 

672 counter: Counter[str] = Counter() 

673 assigned: Dict[str, Set[str]] = defaultdict(set) 

674 

675 # Go through the missing dimensions and associate the 

676 # given names with records within those dimensions 

677 for dimensionName in candidateDimensions: 

678 dimension = self.registry.dimensions.getStaticDimensions()[dimensionName] 

679 fields = dimension.metadata.names | dimension.uniqueKeys.names 

680 for field in not_dimensions: 

681 if field in fields: 

682 guessedAssociation[dimensionName][field] = not_dimensions[field] 

683 counter[dimensionName] += 1 

684 assigned[field].add(dimensionName) 

685 

686 # There is a chance we have allocated a single dataId item 

687 # to multiple dimensions. Need to decide which should be retained. 

688 # For now assume that the most popular alternative wins. 

689 # This means that day_obs with seq_num will result in 

690 # exposure.day_obs and not visit.day_obs 

691 # Also prefer an explicitly missing dimension over an inferred 

692 # temporal dimension. 

693 for fieldName, assignedDimensions in assigned.items(): 

694 if len(assignedDimensions) > 1: 

695 # Pick the most popular (preferring mandatory dimensions) 

696 requiredButMissing = assignedDimensions.intersection(missingDimensions) 

697 if requiredButMissing: 

698 candidateDimensions = requiredButMissing 

699 else: 

700 candidateDimensions = assignedDimensions 

701 

702 # Select the relevant items and get a new restricted 

703 # counter. 

704 theseCounts = {k: v for k, v in counter.items() if k in candidateDimensions} 

705 duplicatesCounter: Counter[str] = Counter() 

706 duplicatesCounter.update(theseCounts) 

707 

708 # Choose the most common. If they are equally common 

709 # we will pick the one that was found first. 

710 # Returns a list of tuples 

711 selected = duplicatesCounter.most_common(1)[0][0] 

712 

713 log.debug("Ambiguous dataId entry '%s' associated with multiple dimensions: %s." 

714 " Removed ambiguity by choosing dimension %s.", 

715 fieldName, ", ".join(assignedDimensions), selected) 

716 

717 for candidateDimension in assignedDimensions: 

718 if candidateDimension != selected: 

719 del guessedAssociation[candidateDimension][fieldName] 

720 

721 # Update the record look up dict with the new associations 

722 for dimensionName, values in guessedAssociation.items(): 

723 if values: # A dict might now be empty 

724 log.debug("Assigned non-dimension dataId keys to dimension %s: %s", 

725 dimensionName, values) 

726 byRecord[dimensionName].update(values) 

727 

728 if byRecord: 

729 # Some record specifiers were found so we need to convert 

730 # them to the Id form 

731 for dimensionName, values in byRecord.items(): 

732 if dimensionName in newDataId: 

733 log.warning("DataId specified explicit %s dimension value of %s in addition to" 

734 " general record specifiers for it of %s. Ignoring record information.", 

735 dimensionName, newDataId[dimensionName], str(values)) 

736 continue 

737 

738 # Build up a WHERE expression -- use single quotes 

739 def quote(s: Any) -> str: 

740 if isinstance(s, str): 

741 return f"'{s}'" 

742 else: 

743 return s 

744 

745 where = " AND ".join(f"{dimensionName}.{k} = {quote(v)}" 

746 for k, v in values.items()) 

747 

748 # Hopefully we get a single record that matches 

749 records = set(self.registry.queryDimensionRecords(dimensionName, dataId=newDataId, 

750 where=where, **kwds)) 

751 

752 if len(records) != 1: 

753 if len(records) > 1: 

754 log.debug("Received %d records from constraints of %s", len(records), str(values)) 

755 for r in records: 

756 log.debug("- %s", str(r)) 

757 raise RuntimeError(f"DataId specification for dimension {dimensionName} is not" 

758 f" uniquely constrained to a single dataset by {values}." 

759 f" Got {len(records)} results.") 

760 raise RuntimeError(f"DataId specification for dimension {dimensionName} matched no" 

761 f" records when constrained by {values}") 

762 

763 # Get the primary key from the real dimension object 

764 dimension = self.registry.dimensions.getStaticDimensions()[dimensionName] 

765 if not isinstance(dimension, Dimension): 

766 raise RuntimeError( 

767 f"{dimension.name} is not a true dimension, and cannot be used in data IDs." 

768 ) 

769 newDataId[dimensionName] = getattr(records.pop(), dimension.primaryKey.name) 

770 

771 # We have modified the dataId so need to switch to it 

772 dataId = newDataId 

773 

774 if datasetType.isCalibration(): 

775 # Because this is a calibration dataset, first try to make a 

776 # standardize the data ID without restricting the dimensions to 

777 # those of the dataset type requested, because there may be extra 

778 # dimensions that provide temporal information for a validity-range 

779 # lookup. 

780 dataId = DataCoordinate.standardize(dataId, universe=self.registry.dimensions, 

781 defaults=self.registry.defaults.dataId, **kwds) 

782 if dataId.graph.temporal: 

783 dataId = self.registry.expandDataId(dataId) 

784 timespan = dataId.timespan 

785 else: 

786 # Standardize the data ID to just the dimensions of the dataset 

787 # type instead of letting registry.findDataset do it, so we get the 

788 # result even if no dataset is found. 

789 dataId = DataCoordinate.standardize(dataId, graph=datasetType.dimensions, 

790 defaults=self.registry.defaults.dataId, **kwds) 

791 # Always lookup the DatasetRef, even if one is given, to ensure it is 

792 # present in the current collection. 

793 ref = self.registry.findDataset(datasetType, dataId, collections=collections, timespan=timespan) 

794 if ref is None: 

795 if allowUnresolved: 

796 return DatasetRef(datasetType, dataId) 

797 else: 

798 if collections is None: 

799 collections = self.registry.defaults.collections 

800 raise LookupError(f"Dataset {datasetType.name} with data ID {dataId} " 

801 f"could not be found in collections {collections}.") 

802 if idNumber is not None and idNumber != ref.id: 

803 if collections is None: 

804 collections = self.registry.defaults.collections 

805 raise ValueError(f"DatasetRef.id provided ({idNumber}) does not match " 

806 f"id ({ref.id}) in registry in collections {collections}.") 

807 return ref 

808 

809 @transactional 

810 def put(self, obj: Any, datasetRefOrType: Union[DatasetRef, DatasetType, str], 

811 dataId: Optional[DataId] = None, *, 

812 run: Optional[str] = None, 

813 **kwds: Any) -> DatasetRef: 

814 """Store and register a dataset. 

815 

816 Parameters 

817 ---------- 

818 obj : `object` 

819 The dataset. 

820 datasetRefOrType : `DatasetRef`, `DatasetType`, or `str` 

821 When `DatasetRef` is provided, ``dataId`` should be `None`. 

822 Otherwise the `DatasetType` or name thereof. 

823 dataId : `dict` or `DataCoordinate` 

824 A `dict` of `Dimension` link name, value pairs that label the 

825 `DatasetRef` within a Collection. When `None`, a `DatasetRef` 

826 should be provided as the second argument. 

827 run : `str`, optional 

828 The name of the run the dataset should be added to, overriding 

829 ``self.run``. 

830 kwds 

831 Additional keyword arguments used to augment or construct a 

832 `DataCoordinate`. See `DataCoordinate.standardize` 

833 parameters. 

834 

835 Returns 

836 ------- 

837 ref : `DatasetRef` 

838 A reference to the stored dataset, updated with the correct id if 

839 given. 

840 

841 Raises 

842 ------ 

843 TypeError 

844 Raised if the butler is read-only or if no run has been provided. 

845 """ 

846 log.debug("Butler put: %s, dataId=%s, run=%s", datasetRefOrType, dataId, run) 

847 if not self.isWriteable(): 

848 raise TypeError("Butler is read-only.") 

849 datasetType, dataId = self._standardizeArgs(datasetRefOrType, dataId, **kwds) 

850 if isinstance(datasetRefOrType, DatasetRef) and datasetRefOrType.id is not None: 

851 raise ValueError("DatasetRef must not be in registry, must have None id") 

852 

853 # Add Registry Dataset entry. 

854 dataId = self.registry.expandDataId(dataId, graph=datasetType.dimensions, **kwds) 

855 ref, = self.registry.insertDatasets(datasetType, run=run, dataIds=[dataId]) 

856 

857 # Add Datastore entry. 

858 self.datastore.put(obj, ref) 

859 

860 return ref 

861 

862 def getDirect(self, ref: DatasetRef, *, parameters: Optional[Dict[str, Any]] = None) -> Any: 

863 """Retrieve a stored dataset. 

864 

865 Unlike `Butler.get`, this method allows datasets outside the Butler's 

866 collection to be read as long as the `DatasetRef` that identifies them 

867 can be obtained separately. 

868 

869 Parameters 

870 ---------- 

871 ref : `DatasetRef` 

872 Resolved reference to an already stored dataset. 

873 parameters : `dict` 

874 Additional StorageClass-defined options to control reading, 

875 typically used to efficiently read only a subset of the dataset. 

876 

877 Returns 

878 ------- 

879 obj : `object` 

880 The dataset. 

881 """ 

882 return self.datastore.get(ref, parameters=parameters) 

883 

884 def getDirectDeferred(self, ref: DatasetRef, *, 

885 parameters: Union[dict, None] = None) -> DeferredDatasetHandle: 

886 """Create a `DeferredDatasetHandle` which can later retrieve a dataset, 

887 from a resolved `DatasetRef`. 

888 

889 Parameters 

890 ---------- 

891 ref : `DatasetRef` 

892 Resolved reference to an already stored dataset. 

893 parameters : `dict` 

894 Additional StorageClass-defined options to control reading, 

895 typically used to efficiently read only a subset of the dataset. 

896 

897 Returns 

898 ------- 

899 obj : `DeferredDatasetHandle` 

900 A handle which can be used to retrieve a dataset at a later time. 

901 

902 Raises 

903 ------ 

904 AmbiguousDatasetError 

905 Raised if ``ref.id is None``, i.e. the reference is unresolved. 

906 """ 

907 if ref.id is None: 

908 raise AmbiguousDatasetError( 

909 f"Dataset of type {ref.datasetType.name} with data ID {ref.dataId} is not resolved." 

910 ) 

911 return DeferredDatasetHandle(butler=self, ref=ref, parameters=parameters) 

912 

913 def getDeferred(self, datasetRefOrType: Union[DatasetRef, DatasetType, str], 

914 dataId: Optional[DataId] = None, *, 

915 parameters: Union[dict, None] = None, 

916 collections: Any = None, 

917 **kwds: Any) -> DeferredDatasetHandle: 

918 """Create a `DeferredDatasetHandle` which can later retrieve a dataset, 

919 after an immediate registry lookup. 

920 

921 Parameters 

922 ---------- 

923 datasetRefOrType : `DatasetRef`, `DatasetType`, or `str` 

924 When `DatasetRef` the `dataId` should be `None`. 

925 Otherwise the `DatasetType` or name thereof. 

926 dataId : `dict` or `DataCoordinate`, optional 

927 A `dict` of `Dimension` link name, value pairs that label the 

928 `DatasetRef` within a Collection. When `None`, a `DatasetRef` 

929 should be provided as the first argument. 

930 parameters : `dict` 

931 Additional StorageClass-defined options to control reading, 

932 typically used to efficiently read only a subset of the dataset. 

933 collections : Any, optional 

934 Collections to be searched, overriding ``self.collections``. 

935 Can be any of the types supported by the ``collections`` argument 

936 to butler construction. 

937 kwds 

938 Additional keyword arguments used to augment or construct a 

939 `DataId`. See `DataId` parameters. 

940 

941 Returns 

942 ------- 

943 obj : `DeferredDatasetHandle` 

944 A handle which can be used to retrieve a dataset at a later time. 

945 

946 Raises 

947 ------ 

948 LookupError 

949 Raised if no matching dataset exists in the `Registry` (and 

950 ``allowUnresolved is False``). 

951 ValueError 

952 Raised if a resolved `DatasetRef` was passed as an input, but it 

953 differs from the one found in the registry. 

954 TypeError 

955 Raised if no collections were provided. 

956 """ 

957 ref = self._findDatasetRef(datasetRefOrType, dataId, collections=collections, **kwds) 

958 return DeferredDatasetHandle(butler=self, ref=ref, parameters=parameters) 

959 

960 def get(self, datasetRefOrType: Union[DatasetRef, DatasetType, str], 

961 dataId: Optional[DataId] = None, *, 

962 parameters: Optional[Dict[str, Any]] = None, 

963 collections: Any = None, 

964 **kwds: Any) -> Any: 

965 """Retrieve a stored dataset. 

966 

967 Parameters 

968 ---------- 

969 datasetRefOrType : `DatasetRef`, `DatasetType`, or `str` 

970 When `DatasetRef` the `dataId` should be `None`. 

971 Otherwise the `DatasetType` or name thereof. 

972 dataId : `dict` or `DataCoordinate` 

973 A `dict` of `Dimension` link name, value pairs that label the 

974 `DatasetRef` within a Collection. When `None`, a `DatasetRef` 

975 should be provided as the first argument. 

976 parameters : `dict` 

977 Additional StorageClass-defined options to control reading, 

978 typically used to efficiently read only a subset of the dataset. 

979 collections : Any, optional 

980 Collections to be searched, overriding ``self.collections``. 

981 Can be any of the types supported by the ``collections`` argument 

982 to butler construction. 

983 kwds 

984 Additional keyword arguments used to augment or construct a 

985 `DataCoordinate`. See `DataCoordinate.standardize` 

986 parameters. 

987 

988 Returns 

989 ------- 

990 obj : `object` 

991 The dataset. 

992 

993 Raises 

994 ------ 

995 ValueError 

996 Raised if a resolved `DatasetRef` was passed as an input, but it 

997 differs from the one found in the registry. 

998 LookupError 

999 Raised if no matching dataset exists in the `Registry`. 

1000 TypeError 

1001 Raised if no collections were provided. 

1002 

1003 Notes 

1004 ----- 

1005 When looking up datasets in a `~CollectionType.CALIBRATION` collection, 

1006 this method requires that the given data ID include temporal dimensions 

1007 beyond the dimensions of the dataset type itself, in order to find the 

1008 dataset with the appropriate validity range. For example, a "bias" 

1009 dataset with native dimensions ``{instrument, detector}`` could be 

1010 fetched with a ``{instrument, detector, exposure}`` data ID, because 

1011 ``exposure`` is a temporal dimension. 

1012 """ 

1013 log.debug("Butler get: %s, dataId=%s, parameters=%s", datasetRefOrType, dataId, parameters) 

1014 ref = self._findDatasetRef(datasetRefOrType, dataId, collections=collections, **kwds) 

1015 return self.getDirect(ref, parameters=parameters) 

1016 

1017 def getURIs(self, datasetRefOrType: Union[DatasetRef, DatasetType, str], 

1018 dataId: Optional[DataId] = None, *, 

1019 predict: bool = False, 

1020 collections: Any = None, 

1021 run: Optional[str] = None, 

1022 **kwds: Any) -> Tuple[Optional[ButlerURI], Dict[str, ButlerURI]]: 

1023 """Returns the URIs associated with the dataset. 

1024 

1025 Parameters 

1026 ---------- 

1027 datasetRefOrType : `DatasetRef`, `DatasetType`, or `str` 

1028 When `DatasetRef` the `dataId` should be `None`. 

1029 Otherwise the `DatasetType` or name thereof. 

1030 dataId : `dict` or `DataCoordinate` 

1031 A `dict` of `Dimension` link name, value pairs that label the 

1032 `DatasetRef` within a Collection. When `None`, a `DatasetRef` 

1033 should be provided as the first argument. 

1034 predict : `bool` 

1035 If `True`, allow URIs to be returned of datasets that have not 

1036 been written. 

1037 collections : Any, optional 

1038 Collections to be searched, overriding ``self.collections``. 

1039 Can be any of the types supported by the ``collections`` argument 

1040 to butler construction. 

1041 run : `str`, optional 

1042 Run to use for predictions, overriding ``self.run``. 

1043 kwds 

1044 Additional keyword arguments used to augment or construct a 

1045 `DataCoordinate`. See `DataCoordinate.standardize` 

1046 parameters. 

1047 

1048 Returns 

1049 ------- 

1050 primary : `ButlerURI` 

1051 The URI to the primary artifact associated with this dataset. 

1052 If the dataset was disassembled within the datastore this 

1053 may be `None`. 

1054 components : `dict` 

1055 URIs to any components associated with the dataset artifact. 

1056 Can be empty if there are no components. 

1057 """ 

1058 ref = self._findDatasetRef(datasetRefOrType, dataId, allowUnresolved=predict, 

1059 collections=collections, **kwds) 

1060 if ref.id is None: # only possible if predict is True 

1061 if run is None: 

1062 run = self.run 

1063 if run is None: 

1064 raise TypeError("Cannot predict location with run=None.") 

1065 # Lie about ID, because we can't guess it, and only 

1066 # Datastore.getURIs() will ever see it (and it doesn't use it). 

1067 ref = ref.resolved(id=0, run=run) 

1068 return self.datastore.getURIs(ref, predict) 

1069 

1070 def getURI(self, datasetRefOrType: Union[DatasetRef, DatasetType, str], 

1071 dataId: Optional[DataId] = None, *, 

1072 predict: bool = False, 

1073 collections: Any = None, 

1074 run: Optional[str] = None, 

1075 **kwds: Any) -> ButlerURI: 

1076 """Return the URI to the Dataset. 

1077 

1078 Parameters 

1079 ---------- 

1080 datasetRefOrType : `DatasetRef`, `DatasetType`, or `str` 

1081 When `DatasetRef` the `dataId` should be `None`. 

1082 Otherwise the `DatasetType` or name thereof. 

1083 dataId : `dict` or `DataCoordinate` 

1084 A `dict` of `Dimension` link name, value pairs that label the 

1085 `DatasetRef` within a Collection. When `None`, a `DatasetRef` 

1086 should be provided as the first argument. 

1087 predict : `bool` 

1088 If `True`, allow URIs to be returned of datasets that have not 

1089 been written. 

1090 collections : Any, optional 

1091 Collections to be searched, overriding ``self.collections``. 

1092 Can be any of the types supported by the ``collections`` argument 

1093 to butler construction. 

1094 run : `str`, optional 

1095 Run to use for predictions, overriding ``self.run``. 

1096 kwds 

1097 Additional keyword arguments used to augment or construct a 

1098 `DataCoordinate`. See `DataCoordinate.standardize` 

1099 parameters. 

1100 

1101 Returns 

1102 ------- 

1103 uri : `ButlerURI` 

1104 URI pointing to the Dataset within the datastore. If the 

1105 Dataset does not exist in the datastore, and if ``predict`` is 

1106 `True`, the URI will be a prediction and will include a URI 

1107 fragment "#predicted". 

1108 If the datastore does not have entities that relate well 

1109 to the concept of a URI the returned URI string will be 

1110 descriptive. The returned URI is not guaranteed to be obtainable. 

1111 

1112 Raises 

1113 ------ 

1114 LookupError 

1115 A URI has been requested for a dataset that does not exist and 

1116 guessing is not allowed. 

1117 ValueError 

1118 Raised if a resolved `DatasetRef` was passed as an input, but it 

1119 differs from the one found in the registry. 

1120 TypeError 

1121 Raised if no collections were provided. 

1122 RuntimeError 

1123 Raised if a URI is requested for a dataset that consists of 

1124 multiple artifacts. 

1125 """ 

1126 primary, components = self.getURIs(datasetRefOrType, dataId=dataId, predict=predict, 

1127 collections=collections, run=run, **kwds) 

1128 

1129 if primary is None or components: 

1130 raise RuntimeError(f"Dataset ({datasetRefOrType}) includes distinct URIs for components. " 

1131 "Use Butler.getURIs() instead.") 

1132 return primary 

1133 

1134 def retrieveArtifacts(self, refs: Iterable[DatasetRef], 

1135 destination: Union[str, ButlerURI], transfer: str = "auto", 

1136 preserve_path: bool = True, 

1137 overwrite: bool = False) -> List[ButlerURI]: 

1138 """Retrieve the artifacts associated with the supplied refs. 

1139 

1140 Parameters 

1141 ---------- 

1142 refs : iterable of `DatasetRef` 

1143 The datasets for which artifacts are to be retrieved. 

1144 A single ref can result in multiple artifacts. The refs must 

1145 be resolved. 

1146 destination : `ButlerURI` or `str` 

1147 Location to write the artifacts. 

1148 transfer : `str`, optional 

1149 Method to use to transfer the artifacts. Must be one of the options 

1150 supported by `ButlerURI.transfer_from()`. "move" is not allowed. 

1151 preserve_path : `bool`, optional 

1152 If `True` the full path of the artifact within the datastore 

1153 is preserved. If `False` the final file component of the path 

1154 is used. 

1155 overwrite : `bool`, optional 

1156 If `True` allow transfers to overwrite existing files at the 

1157 destination. 

1158 

1159 Returns 

1160 ------- 

1161 targets : `list` of `ButlerURI` 

1162 URIs of file artifacts in destination location. Order is not 

1163 preserved. 

1164 

1165 Notes 

1166 ----- 

1167 For non-file datastores the artifacts written to the destination 

1168 may not match the representation inside the datastore. For example 

1169 a hierarchical data structure in a NoSQL database may well be stored 

1170 as a JSON file. 

1171 """ 

1172 return self.datastore.retrieveArtifacts(refs, ButlerURI(destination), transfer=transfer, 

1173 preserve_path=preserve_path, overwrite=overwrite) 

1174 

1175 def datasetExists(self, datasetRefOrType: Union[DatasetRef, DatasetType, str], 

1176 dataId: Optional[DataId] = None, *, 

1177 collections: Any = None, 

1178 **kwds: Any) -> bool: 

1179 """Return True if the Dataset is actually present in the Datastore. 

1180 

1181 Parameters 

1182 ---------- 

1183 datasetRefOrType : `DatasetRef`, `DatasetType`, or `str` 

1184 When `DatasetRef` the `dataId` should be `None`. 

1185 Otherwise the `DatasetType` or name thereof. 

1186 dataId : `dict` or `DataCoordinate` 

1187 A `dict` of `Dimension` link name, value pairs that label the 

1188 `DatasetRef` within a Collection. When `None`, a `DatasetRef` 

1189 should be provided as the first argument. 

1190 collections : Any, optional 

1191 Collections to be searched, overriding ``self.collections``. 

1192 Can be any of the types supported by the ``collections`` argument 

1193 to butler construction. 

1194 kwds 

1195 Additional keyword arguments used to augment or construct a 

1196 `DataCoordinate`. See `DataCoordinate.standardize` 

1197 parameters. 

1198 

1199 Raises 

1200 ------ 

1201 LookupError 

1202 Raised if the dataset is not even present in the Registry. 

1203 ValueError 

1204 Raised if a resolved `DatasetRef` was passed as an input, but it 

1205 differs from the one found in the registry. 

1206 TypeError 

1207 Raised if no collections were provided. 

1208 """ 

1209 ref = self._findDatasetRef(datasetRefOrType, dataId, collections=collections, **kwds) 

1210 return self.datastore.exists(ref) 

1211 

1212 def removeRuns(self, names: Iterable[str], unstore: bool = True) -> None: 

1213 """Remove one or more `~CollectionType.RUN` collections and the 

1214 datasets within them. 

1215 

1216 Parameters 

1217 ---------- 

1218 names : `Iterable` [ `str` ] 

1219 The names of the collections to remove. 

1220 unstore : `bool`, optional 

1221 If `True` (default), delete datasets from all datastores in which 

1222 they are present, and attempt to rollback the registry deletions if 

1223 datastore deletions fail (which may not always be possible). If 

1224 `False`, datastore records for these datasets are still removed, 

1225 but any artifacts (e.g. files) will not be. 

1226 

1227 Raises 

1228 ------ 

1229 TypeError 

1230 Raised if one or more collections are not of type 

1231 `~CollectionType.RUN`. 

1232 """ 

1233 if not self.isWriteable(): 

1234 raise TypeError("Butler is read-only.") 

1235 names = list(names) 

1236 refs: List[DatasetRef] = [] 

1237 for name in names: 

1238 collectionType = self.registry.getCollectionType(name) 

1239 if collectionType is not CollectionType.RUN: 

1240 raise TypeError(f"The collection type of '{name}' is {collectionType.name}, not RUN.") 

1241 refs.extend(self.registry.queryDatasets(..., collections=name, findFirst=True)) 

1242 with self.registry.transaction(): 

1243 if unstore: 

1244 self.datastore.trash(refs) 

1245 else: 

1246 self.datastore.forget(refs) 

1247 for name in names: 

1248 self.registry.removeCollection(name) 

1249 if unstore: 

1250 # Point of no return for removing artifacts 

1251 self.datastore.emptyTrash() 

1252 

1253 def pruneCollection(self, name: str, purge: bool = False, unstore: bool = False, 

1254 unlink: Optional[List[str]] = None) -> None: 

1255 """Remove a collection and possibly prune datasets within it. 

1256 

1257 Parameters 

1258 ---------- 

1259 name : `str` 

1260 Name of the collection to remove. If this is a 

1261 `~CollectionType.TAGGED` or `~CollectionType.CHAINED` collection, 

1262 datasets within the collection are not modified unless ``unstore`` 

1263 is `True`. If this is a `~CollectionType.RUN` collection, 

1264 ``purge`` and ``unstore`` must be `True`, and all datasets in it 

1265 are fully removed from the data repository. 

1266 purge : `bool`, optional 

1267 If `True`, permit `~CollectionType.RUN` collections to be removed, 

1268 fully removing datasets within them. Requires ``unstore=True`` as 

1269 well as an added precaution against accidental deletion. Must be 

1270 `False` (default) if the collection is not a ``RUN``. 

1271 unstore: `bool`, optional 

1272 If `True`, remove all datasets in the collection from all 

1273 datastores in which they appear. 

1274 unlink: `list` [`str`], optional 

1275 Before removing the given `collection` unlink it from from these 

1276 parent collections. 

1277 

1278 Raises 

1279 ------ 

1280 TypeError 

1281 Raised if the butler is read-only or arguments are mutually 

1282 inconsistent. 

1283 """ 

1284 # See pruneDatasets comments for more information about the logic here; 

1285 # the cases are almost the same, but here we can rely on Registry to 

1286 # take care everything but Datastore deletion when we remove the 

1287 # collection. 

1288 if not self.isWriteable(): 

1289 raise TypeError("Butler is read-only.") 

1290 collectionType = self.registry.getCollectionType(name) 

1291 if purge and not unstore: 

1292 raise PurgeWithoutUnstorePruneCollectionsError() 

1293 if collectionType is CollectionType.RUN and not purge: 

1294 raise RunWithoutPurgePruneCollectionsError(collectionType) 

1295 if collectionType is not CollectionType.RUN and purge: 

1296 raise PurgeUnsupportedPruneCollectionsError(collectionType) 

1297 

1298 def remove(child: str, parent: str) -> None: 

1299 """Remove a child collection from a parent collection.""" 

1300 # Remove child from parent. 

1301 chain = list(self.registry.getCollectionChain(parent)) 

1302 try: 

1303 chain.remove(name) 

1304 except ValueError as e: 

1305 raise RuntimeError(f"{name} is not a child of {parent}") from e 

1306 self.registry.setCollectionChain(parent, chain) 

1307 

1308 with self.registry.transaction(): 

1309 if (unlink): 

1310 for parent in unlink: 

1311 remove(name, parent) 

1312 if unstore: 

1313 refs = self.registry.queryDatasets(..., collections=name, findFirst=True) 

1314 self.datastore.trash(refs) 

1315 self.registry.removeCollection(name) 

1316 

1317 if unstore: 

1318 # Point of no return for removing artifacts 

1319 self.datastore.emptyTrash() 

1320 

1321 def pruneDatasets(self, refs: Iterable[DatasetRef], *, 

1322 disassociate: bool = True, 

1323 unstore: bool = False, 

1324 tags: Iterable[str] = (), 

1325 purge: bool = False, 

1326 run: Optional[str] = None) -> None: 

1327 """Remove one or more datasets from a collection and/or storage. 

1328 

1329 Parameters 

1330 ---------- 

1331 refs : `~collections.abc.Iterable` of `DatasetRef` 

1332 Datasets to prune. These must be "resolved" references (not just 

1333 a `DatasetType` and data ID). 

1334 disassociate : `bool`, optional 

1335 Disassociate pruned datasets from ``tags``, or from all collections 

1336 if ``purge=True``. 

1337 unstore : `bool`, optional 

1338 If `True` (`False` is default) remove these datasets from all 

1339 datastores known to this butler. Note that this will make it 

1340 impossible to retrieve these datasets even via other collections. 

1341 Datasets that are already not stored are ignored by this option. 

1342 tags : `Iterable` [ `str` ], optional 

1343 `~CollectionType.TAGGED` collections to disassociate the datasets 

1344 from. Ignored if ``disassociate`` is `False` or ``purge`` is 

1345 `True`. 

1346 purge : `bool`, optional 

1347 If `True` (`False` is default), completely remove the dataset from 

1348 the `Registry`. To prevent accidental deletions, ``purge`` may 

1349 only be `True` if all of the following conditions are met: 

1350 

1351 - All given datasets are in the given run. 

1352 - ``disassociate`` is `True`; 

1353 - ``unstore`` is `True`. 

1354 

1355 This mode may remove provenance information from datasets other 

1356 than those provided, and should be used with extreme care. 

1357 

1358 Raises 

1359 ------ 

1360 TypeError 

1361 Raised if the butler is read-only, if no collection was provided, 

1362 or the conditions for ``purge=True`` were not met. 

1363 """ 

1364 if not self.isWriteable(): 

1365 raise TypeError("Butler is read-only.") 

1366 if purge: 

1367 if not disassociate: 

1368 raise TypeError("Cannot pass purge=True without disassociate=True.") 

1369 if not unstore: 

1370 raise TypeError("Cannot pass purge=True without unstore=True.") 

1371 elif disassociate: 

1372 tags = tuple(tags) 

1373 if not tags: 

1374 raise TypeError("No tags provided but disassociate=True.") 

1375 for tag in tags: 

1376 collectionType = self.registry.getCollectionType(tag) 

1377 if collectionType is not CollectionType.TAGGED: 

1378 raise TypeError(f"Cannot disassociate from collection '{tag}' " 

1379 f"of non-TAGGED type {collectionType.name}.") 

1380 # Transform possibly-single-pass iterable into something we can iterate 

1381 # over multiple times. 

1382 refs = list(refs) 

1383 # Pruning a component of a DatasetRef makes no sense since registry 

1384 # doesn't know about components and datastore might not store 

1385 # components in a separate file 

1386 for ref in refs: 

1387 if ref.datasetType.component(): 

1388 raise ValueError(f"Can not prune a component of a dataset (ref={ref})") 

1389 # We don't need an unreliable Datastore transaction for this, because 

1390 # we've been extra careful to ensure that Datastore.trash only involves 

1391 # mutating the Registry (it can _look_ at Datastore-specific things, 

1392 # but shouldn't change them), and hence all operations here are 

1393 # Registry operations. 

1394 with self.registry.transaction(): 

1395 if unstore: 

1396 self.datastore.trash(refs) 

1397 if purge: 

1398 self.registry.removeDatasets(refs) 

1399 elif disassociate: 

1400 assert tags, "Guaranteed by earlier logic in this function." 

1401 for tag in tags: 

1402 self.registry.disassociate(tag, refs) 

1403 # We've exited the Registry transaction, and apparently committed. 

1404 # (if there was an exception, everything rolled back, and it's as if 

1405 # nothing happened - and we never get here). 

1406 # Datastore artifacts are not yet gone, but they're clearly marked 

1407 # as trash, so if we fail to delete now because of (e.g.) filesystem 

1408 # problems we can try again later, and if manual administrative 

1409 # intervention is required, it's pretty clear what that should entail: 

1410 # deleting everything on disk and in private Datastore tables that is 

1411 # in the dataset_location_trash table. 

1412 if unstore: 

1413 # Point of no return for removing artifacts 

1414 self.datastore.emptyTrash() 

1415 

1416 @transactional 

1417 def ingest(self, *datasets: FileDataset, transfer: Optional[str] = "auto", run: Optional[str] = None, 

1418 idGenerationMode: DatasetIdGenEnum = DatasetIdGenEnum.UNIQUE, 

1419 ) -> None: 

1420 """Store and register one or more datasets that already exist on disk. 

1421 

1422 Parameters 

1423 ---------- 

1424 datasets : `FileDataset` 

1425 Each positional argument is a struct containing information about 

1426 a file to be ingested, including its path (either absolute or 

1427 relative to the datastore root, if applicable), a `DatasetRef`, 

1428 and optionally a formatter class or its fully-qualified string 

1429 name. If a formatter is not provided, the formatter that would be 

1430 used for `put` is assumed. On successful return, all 

1431 `FileDataset.ref` attributes will have their `DatasetRef.id` 

1432 attribute populated and all `FileDataset.formatter` attributes will 

1433 be set to the formatter class used. `FileDataset.path` attributes 

1434 may be modified to put paths in whatever the datastore considers a 

1435 standardized form. 

1436 transfer : `str`, optional 

1437 If not `None`, must be one of 'auto', 'move', 'copy', 'direct', 

1438 'hardlink', 'relsymlink' or 'symlink', indicating how to transfer 

1439 the file. 

1440 run : `str`, optional 

1441 The name of the run ingested datasets should be added to, 

1442 overriding ``self.run``. 

1443 idGenerationMode : `DatasetIdGenEnum`, optional 

1444 Specifies option for generating dataset IDs. By default unique IDs 

1445 are generated for each inserted dataset. 

1446 

1447 Raises 

1448 ------ 

1449 TypeError 

1450 Raised if the butler is read-only or if no run was provided. 

1451 NotImplementedError 

1452 Raised if the `Datastore` does not support the given transfer mode. 

1453 DatasetTypeNotSupportedError 

1454 Raised if one or more files to be ingested have a dataset type that 

1455 is not supported by the `Datastore`.. 

1456 FileNotFoundError 

1457 Raised if one of the given files does not exist. 

1458 FileExistsError 

1459 Raised if transfer is not `None` but the (internal) location the 

1460 file would be moved to is already occupied. 

1461 

1462 Notes 

1463 ----- 

1464 This operation is not fully exception safe: if a database operation 

1465 fails, the given `FileDataset` instances may be only partially updated. 

1466 

1467 It is atomic in terms of database operations (they will either all 

1468 succeed or all fail) providing the database engine implements 

1469 transactions correctly. It will attempt to be atomic in terms of 

1470 filesystem operations as well, but this cannot be implemented 

1471 rigorously for most datastores. 

1472 """ 

1473 if not self.isWriteable(): 

1474 raise TypeError("Butler is read-only.") 

1475 progress = Progress("lsst.daf.butler.Butler.ingest", level=logging.DEBUG) 

1476 # Reorganize the inputs so they're grouped by DatasetType and then 

1477 # data ID. We also include a list of DatasetRefs for each FileDataset 

1478 # to hold the resolved DatasetRefs returned by the Registry, before 

1479 # it's safe to swap them into FileDataset.refs. 

1480 # Some type annotation aliases to make that clearer: 

1481 GroupForType = Dict[DataCoordinate, Tuple[FileDataset, List[DatasetRef]]] 

1482 GroupedData = MutableMapping[DatasetType, GroupForType] 

1483 # The actual data structure: 

1484 groupedData: GroupedData = defaultdict(dict) 

1485 # And the nested loop that populates it: 

1486 for dataset in progress.wrap(datasets, desc="Grouping by dataset type"): 

1487 # This list intentionally shared across the inner loop, since it's 

1488 # associated with `dataset`. 

1489 resolvedRefs: List[DatasetRef] = [] 

1490 for ref in dataset.refs: 

1491 if ref.dataId in groupedData[ref.datasetType]: 

1492 raise ConflictingDefinitionError(f"Ingest conflict. Dataset {dataset.path} has same" 

1493 " DataId as other ingest dataset" 

1494 f" {groupedData[ref.datasetType][ref.dataId][0].path} " 

1495 f" ({ref.dataId})") 

1496 groupedData[ref.datasetType][ref.dataId] = (dataset, resolvedRefs) 

1497 

1498 # Now we can bulk-insert into Registry for each DatasetType. 

1499 allResolvedRefs: List[DatasetRef] = [] 

1500 for datasetType, groupForType in progress.iter_item_chunks(groupedData.items(), 

1501 desc="Bulk-inserting datasets by type"): 

1502 refs = self.registry.insertDatasets( 

1503 datasetType, 

1504 dataIds=groupForType.keys(), 

1505 run=run, 

1506 expand=self.datastore.needs_expanded_data_ids(transfer, datasetType), 

1507 idGenerationMode=idGenerationMode, 

1508 ) 

1509 # Append those resolved DatasetRefs to the new lists we set up for 

1510 # them. 

1511 for ref, (_, resolvedRefs) in zip(refs, groupForType.values()): 

1512 resolvedRefs.append(ref) 

1513 

1514 # Go back to the original FileDatasets to replace their refs with the 

1515 # new resolved ones, and also build a big list of all refs. 

1516 allResolvedRefs = [] 

1517 for groupForType in progress.iter_chunks(groupedData.values(), 

1518 desc="Reassociating resolved dataset refs with files"): 

1519 for dataset, resolvedRefs in groupForType.values(): 

1520 dataset.refs = resolvedRefs 

1521 allResolvedRefs.extend(resolvedRefs) 

1522 

1523 # Bulk-insert everything into Datastore. 

1524 self.datastore.ingest(*datasets, transfer=transfer) 

1525 

1526 @contextlib.contextmanager 

1527 def export(self, *, directory: Optional[str] = None, 

1528 filename: Optional[str] = None, 

1529 format: Optional[str] = None, 

1530 transfer: Optional[str] = None) -> Iterator[RepoExportContext]: 

1531 """Export datasets from the repository represented by this `Butler`. 

1532 

1533 This method is a context manager that returns a helper object 

1534 (`RepoExportContext`) that is used to indicate what information from 

1535 the repository should be exported. 

1536 

1537 Parameters 

1538 ---------- 

1539 directory : `str`, optional 

1540 Directory dataset files should be written to if ``transfer`` is not 

1541 `None`. 

1542 filename : `str`, optional 

1543 Name for the file that will include database information associated 

1544 with the exported datasets. If this is not an absolute path and 

1545 ``directory`` is not `None`, it will be written to ``directory`` 

1546 instead of the current working directory. Defaults to 

1547 "export.{format}". 

1548 format : `str`, optional 

1549 File format for the database information file. If `None`, the 

1550 extension of ``filename`` will be used. 

1551 transfer : `str`, optional 

1552 Transfer mode passed to `Datastore.export`. 

1553 

1554 Raises 

1555 ------ 

1556 TypeError 

1557 Raised if the set of arguments passed is inconsistent. 

1558 

1559 Examples 

1560 -------- 

1561 Typically the `Registry.queryDataIds` and `Registry.queryDatasets` 

1562 methods are used to provide the iterables over data IDs and/or datasets 

1563 to be exported:: 

1564 

1565 with butler.export("exports.yaml") as export: 

1566 # Export all flats, but none of the dimension element rows 

1567 # (i.e. data ID information) associated with them. 

1568 export.saveDatasets(butler.registry.queryDatasets("flat"), 

1569 elements=()) 

1570 # Export all datasets that start with "deepCoadd_" and all of 

1571 # their associated data ID information. 

1572 export.saveDatasets(butler.registry.queryDatasets("deepCoadd_*")) 

1573 """ 

1574 if directory is None and transfer is not None: 

1575 raise TypeError("Cannot transfer without providing a directory.") 

1576 if transfer == "move": 

1577 raise TypeError("Transfer may not be 'move': export is read-only") 

1578 if format is None: 

1579 if filename is None: 

1580 raise TypeError("At least one of 'filename' or 'format' must be provided.") 

1581 else: 

1582 _, format = os.path.splitext(filename) 

1583 elif filename is None: 

1584 filename = f"export.{format}" 

1585 if directory is not None: 

1586 filename = os.path.join(directory, filename) 

1587 BackendClass = getClassOf(self._config["repo_transfer_formats"][format]["export"]) 

1588 with open(filename, 'w') as stream: 

1589 backend = BackendClass(stream) 

1590 try: 

1591 helper = RepoExportContext(self.registry, self.datastore, backend=backend, 

1592 directory=directory, transfer=transfer) 

1593 yield helper 

1594 except BaseException: 

1595 raise 

1596 else: 

1597 helper._finish() 

1598 

1599 def import_(self, *, directory: Optional[str] = None, 

1600 filename: Union[str, TextIO, None] = None, 

1601 format: Optional[str] = None, 

1602 transfer: Optional[str] = None, 

1603 skip_dimensions: Optional[Set] = None, 

1604 idGenerationMode: DatasetIdGenEnum = DatasetIdGenEnum.UNIQUE, 

1605 reuseIds: bool = False) -> None: 

1606 """Import datasets into this repository that were exported from a 

1607 different butler repository via `~lsst.daf.butler.Butler.export`. 

1608 

1609 Parameters 

1610 ---------- 

1611 directory : `str`, optional 

1612 Directory containing dataset files to import from. If `None`, 

1613 ``filename`` and all dataset file paths specified therein must 

1614 be absolute. 

1615 filename : `str` or `TextIO`, optional 

1616 A stream or name of file that contains database information 

1617 associated with the exported datasets, typically generated by 

1618 `~lsst.daf.butler.Butler.export`. If this a string (name) and 

1619 is not an absolute path, does not exist in the current working 

1620 directory, and ``directory`` is not `None`, it is assumed to be in 

1621 ``directory``. Defaults to "export.{format}". 

1622 format : `str`, optional 

1623 File format for ``filename``. If `None`, the extension of 

1624 ``filename`` will be used. 

1625 transfer : `str`, optional 

1626 Transfer mode passed to `~lsst.daf.butler.Datastore.ingest`. 

1627 skip_dimensions : `set`, optional 

1628 Names of dimensions that should be skipped and not imported. 

1629 idGenerationMode : `DatasetIdGenEnum`, optional 

1630 Specifies option for generating dataset IDs when IDs are not 

1631 provided or their type does not match backend type. By default 

1632 unique IDs are generated for each inserted dataset. 

1633 reuseIds : `bool`, optional 

1634 If `True` then forces re-use of imported dataset IDs for integer 

1635 IDs which are normally generated as auto-incremented; exception 

1636 will be raised if imported IDs clash with existing ones. This 

1637 option has no effect on the use of globally-unique IDs which are 

1638 always re-used (or generated if integer IDs are being imported). 

1639 

1640 Raises 

1641 ------ 

1642 TypeError 

1643 Raised if the set of arguments passed is inconsistent, or if the 

1644 butler is read-only. 

1645 """ 

1646 if not self.isWriteable(): 

1647 raise TypeError("Butler is read-only.") 

1648 if format is None: 

1649 if filename is None: 

1650 raise TypeError("At least one of 'filename' or 'format' must be provided.") 

1651 else: 

1652 _, format = os.path.splitext(filename) # type: ignore 

1653 elif filename is None: 

1654 filename = f"export.{format}" 

1655 if isinstance(filename, str) and directory is not None and not os.path.exists(filename): 

1656 filename = os.path.join(directory, filename) 

1657 BackendClass = getClassOf(self._config["repo_transfer_formats"][format]["import"]) 

1658 

1659 def doImport(importStream: TextIO) -> None: 

1660 backend = BackendClass(importStream, self.registry) 

1661 backend.register() 

1662 with self.transaction(): 

1663 backend.load(self.datastore, directory=directory, transfer=transfer, 

1664 skip_dimensions=skip_dimensions, idGenerationMode=idGenerationMode, 

1665 reuseIds=reuseIds) 

1666 

1667 if isinstance(filename, str): 

1668 with open(filename, "r") as stream: 

1669 doImport(stream) 

1670 else: 

1671 doImport(filename) 

1672 

1673 def validateConfiguration(self, logFailures: bool = False, 

1674 datasetTypeNames: Optional[Iterable[str]] = None, 

1675 ignore: Iterable[str] = None) -> None: 

1676 """Validate butler configuration. 

1677 

1678 Checks that each `DatasetType` can be stored in the `Datastore`. 

1679 

1680 Parameters 

1681 ---------- 

1682 logFailures : `bool`, optional 

1683 If `True`, output a log message for every validation error 

1684 detected. 

1685 datasetTypeNames : iterable of `str`, optional 

1686 The `DatasetType` names that should be checked. This allows 

1687 only a subset to be selected. 

1688 ignore : iterable of `str`, optional 

1689 Names of DatasetTypes to skip over. This can be used to skip 

1690 known problems. If a named `DatasetType` corresponds to a 

1691 composite, all components of that `DatasetType` will also be 

1692 ignored. 

1693 

1694 Raises 

1695 ------ 

1696 ButlerValidationError 

1697 Raised if there is some inconsistency with how this Butler 

1698 is configured. 

1699 """ 

1700 if datasetTypeNames: 

1701 datasetTypes = [self.registry.getDatasetType(name) for name in datasetTypeNames] 

1702 else: 

1703 datasetTypes = list(self.registry.queryDatasetTypes()) 

1704 

1705 # filter out anything from the ignore list 

1706 if ignore: 

1707 ignore = set(ignore) 

1708 datasetTypes = [e for e in datasetTypes 

1709 if e.name not in ignore and e.nameAndComponent()[0] not in ignore] 

1710 else: 

1711 ignore = set() 

1712 

1713 # Find all the registered instruments 

1714 instruments = set( 

1715 record.name for record in self.registry.queryDimensionRecords("instrument") 

1716 ) 

1717 

1718 # For each datasetType that has an instrument dimension, create 

1719 # a DatasetRef for each defined instrument 

1720 datasetRefs = [] 

1721 

1722 for datasetType in datasetTypes: 

1723 if "instrument" in datasetType.dimensions: 

1724 for instrument in instruments: 

1725 datasetRef = DatasetRef(datasetType, {"instrument": instrument}, # type: ignore 

1726 conform=False) 

1727 datasetRefs.append(datasetRef) 

1728 

1729 entities: List[Union[DatasetType, DatasetRef]] = [] 

1730 entities.extend(datasetTypes) 

1731 entities.extend(datasetRefs) 

1732 

1733 datastoreErrorStr = None 

1734 try: 

1735 self.datastore.validateConfiguration(entities, logFailures=logFailures) 

1736 except ValidationError as e: 

1737 datastoreErrorStr = str(e) 

1738 

1739 # Also check that the LookupKeys used by the datastores match 

1740 # registry and storage class definitions 

1741 keys = self.datastore.getLookupKeys() 

1742 

1743 failedNames = set() 

1744 failedDataId = set() 

1745 for key in keys: 

1746 if key.name is not None: 

1747 if key.name in ignore: 

1748 continue 

1749 

1750 # skip if specific datasetType names were requested and this 

1751 # name does not match 

1752 if datasetTypeNames and key.name not in datasetTypeNames: 

1753 continue 

1754 

1755 # See if it is a StorageClass or a DatasetType 

1756 if key.name in self.storageClasses: 

1757 pass 

1758 else: 

1759 try: 

1760 self.registry.getDatasetType(key.name) 

1761 except KeyError: 

1762 if logFailures: 

1763 log.critical("Key '%s' does not correspond to a DatasetType or StorageClass", key) 

1764 failedNames.add(key) 

1765 else: 

1766 # Dimensions are checked for consistency when the Butler 

1767 # is created and rendezvoused with a universe. 

1768 pass 

1769 

1770 # Check that the instrument is a valid instrument 

1771 # Currently only support instrument so check for that 

1772 if key.dataId: 

1773 dataIdKeys = set(key.dataId) 

1774 if set(["instrument"]) != dataIdKeys: 

1775 if logFailures: 

1776 log.critical("Key '%s' has unsupported DataId override", key) 

1777 failedDataId.add(key) 

1778 elif key.dataId["instrument"] not in instruments: 

1779 if logFailures: 

1780 log.critical("Key '%s' has unknown instrument", key) 

1781 failedDataId.add(key) 

1782 

1783 messages = [] 

1784 

1785 if datastoreErrorStr: 

1786 messages.append(datastoreErrorStr) 

1787 

1788 for failed, msg in ((failedNames, "Keys without corresponding DatasetType or StorageClass entry: "), 

1789 (failedDataId, "Keys with bad DataId entries: ")): 

1790 if failed: 

1791 msg += ", ".join(str(k) for k in failed) 

1792 messages.append(msg) 

1793 

1794 if messages: 

1795 raise ValidationError(";\n".join(messages)) 

1796 

1797 @property 

1798 def collections(self) -> CollectionSearch: 

1799 """The collections to search by default, in order (`CollectionSearch`). 

1800 

1801 This is an alias for ``self.registry.defaults.collections``. It cannot 

1802 be set directly in isolation, but all defaults may be changed together 

1803 by assigning a new `RegistryDefaults` instance to 

1804 ``self.registry.defaults``. 

1805 """ 

1806 return self.registry.defaults.collections 

1807 

1808 @property 

1809 def run(self) -> Optional[str]: 

1810 """Name of the run this butler writes outputs to by default (`str` or 

1811 `None`). 

1812 

1813 This is an alias for ``self.registry.defaults.run``. It cannot be set 

1814 directly in isolation, but all defaults may be changed together by 

1815 assigning a new `RegistryDefaults` instance to 

1816 ``self.registry.defaults``. 

1817 """ 

1818 return self.registry.defaults.run 

1819 

1820 registry: Registry 

1821 """The object that manages dataset metadata and relationships (`Registry`). 

1822 

1823 Most operations that don't involve reading or writing butler datasets are 

1824 accessible only via `Registry` methods. 

1825 """ 

1826 

1827 datastore: Datastore 

1828 """The object that manages actual dataset storage (`Datastore`). 

1829 

1830 Direct user access to the datastore should rarely be necessary; the primary 

1831 exception is the case where a `Datastore` implementation provides extra 

1832 functionality beyond what the base class defines. 

1833 """ 

1834 

1835 storageClasses: StorageClassFactory 

1836 """An object that maps known storage class names to objects that fully 

1837 describe them (`StorageClassFactory`). 

1838 """